Infinite Latent Feature Selection: A Probabilistic Latent Graph-Based Ranking Approach

被引:168
|
作者
Roffo, Giorgio [1 ]
Melzi, Simone [2 ]
Castellani, Umberto [2 ]
Vinciarelli, Alessandro [1 ]
机构
[1] Univ Glasgow, Glasgow, Lanark, Scotland
[2] Univ Verona, Verona, Italy
基金
英国工程与自然科学研究理事会;
关键词
CANCER; CLASSIFICATION;
D O I
10.1109/ICCV.2017.156
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Feature selection is playing an increasingly significant role with respect to many computer vision applications spanning from object recognition to visual object tracking. However, most of the recent solutions in feature selection are not robust across different and heterogeneous set of data. In this paper, we address this issue proposing a robust probabilistic latent graph-based feature selection algorithm that performs the ranking step while considering all the possible subsets of features, as paths on a graph, bypassing the combinatorial problem analytically. An appealing characteristic of the approach is that it aims to discover an abstraction behind low-level sensory data, that is, relevancy. Relevancy is modelled as a latent variable in a PLSA-inspired generative process that allows the investigation of the importance of a feature when injected into an arbitrary set of cues. The proposed method has been tested on ten diverse benchmarks, and compared against eleven state of the art feature selection methods. Results show that the proposed approach attains the highest performance levels across many different scenarios and difficulties, thereby confirming its strong robustness while setting a new state of the art in feature selection domain.
引用
收藏
页码:1407 / 1415
页数:9
相关论文
共 50 条
  • [1] Infinite Feature Selection: A Graph-based Feature Filtering Approach
    Roffo, Giorgio
    Melzi, Simone
    Castellani, Umberto
    Vinciarelli, Alessandro
    Cristani, Marco
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2021, 43 (12) : 4396 - 4410
  • [2] A Graph-Based Approach to Feature Selection
    Zhang, Zhihong
    Hancock, Edwin R.
    [J]. GRAPH-BASED REPRESENTATIONS IN PATTERN RECOGNITION, 2011, 6658 : 205 - 214
  • [3] Feature grouping and selection: A graph-based approach
    Zheng, Ling
    Chao, Fei
    Mac Parthalain, Neil
    Zhang, Defu
    Shen, Qiang
    [J]. INFORMATION SCIENCES, 2021, 546 : 1256 - 1272
  • [4] Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking
    Hu YaBin
    Ren GuangBo
    Ma Yi
    Yang JunFang
    Wang JianBu
    An JuBai
    Liang Jian
    Ma YuanQing
    Song XiuKai
    [J]. SCIENCE CHINA-TECHNOLOGICAL SCIENCES, 2022, 65 (04) : 759 - 777
  • [5] Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking
    YaBin Hu
    GuangBo Ren
    Yi Ma
    JunFang Yang
    JianBu Wang
    JuBai An
    Jian Liang
    YuanQing Ma
    XiuKai Song
    [J]. Science China Technological Sciences, 2022, 65 : 759 - 777
  • [6] Combining probabilistic ranking and latent semantic indexing for feature identification
    Poshyvanyk, Denys
    Guehenuec, Yann-Gael
    Marcus, Andrian
    Antoniol, Giuliano
    Rajlich, Vaclav
    [J]. 14TH IEEE INTERNATIONAL CONFERENCE ON PROGRAM COMPREHENSION (ICPC 2006), PROCEEDINGS, 2006, : 137 - +
  • [7] Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking
    HU YaBin
    REN GuangBo
    MA Yi
    YANG JunFang
    WANG JianBu
    AN JuBai
    LIANG Jian
    MA YuanQing
    SONG XiuKai
    [J]. Science China Technological Sciences, 2022, 65 (04) : 759 - 777
  • [8] Coastal wetland hyperspectral classification under the collaborative of subspace partition and infinite probabilistic latent graph ranking
    HU YaBin
    REN GuangBo
    MA Yi
    YANG JunFang
    WANG JianBu
    AN JuBai
    LIANG Jian
    MA YuanQing
    SONG XiuKai
    [J]. Science China(Technological Sciences), 2022, (04) - 777
  • [9] Graph-based bootstrapped latent recommendation model
    Wang, Heyong
    Jiang, Guanshang
    Hong, Ming
    Abdalbari, Headar
    [J]. ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS, 2024, 68
  • [10] Graph-Based Feature Selection Approach for Molecular Activity Prediction
    Cerruela-Garcia, Gonzalo
    Manuel Cuevas-Munoz, Jose
    Garcia-Pedrajas, Nicolas
    [J]. JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2022, 62 (07) : 1618 - 1632